Forum Stats

  • 3,826,744 Users
  • 2,260,702 Discussions
  • 7,897,069 Comments

Discussions

Text Search Indexer for Japanese

Hi,
I have been creating OHJ help set in Japanese and having a trouble with a .idx file for full-text searches generated by the OHJ Text Search Indexer.

As the "Using the Text Search Indexer" topic in the ohguide.hs manual instructed, I was able to generate the .idx file. But the number of words and phrases to matche is very limited in comparison with English text.

Is this a normal behaivor, or is there any way I can improve the number of matches?

Thank you for your support,
Rumiko

Comments

  • 3004
    3004 Member Posts: 204,171 Green Ribbon
    Are you using the Japanese Indexer? The Japanese Indexer uses the Java BreakIterator to identify "words" in
    Japanese. Development has also added some additional processing on top of that to handle some additional breaking up and indexing for certain writing systems.

    For details about the Japanese Indexer, see http://otn.oracle.com/ohguide/help/topic?inOHW=false&file=file%3A/u01/webapps/OHW/ohw-app/ohguide/helpsets/ohguide/oha_gen_fts.html&linkHelp=false#Running%20the%20JapaneseIndexer.

    Summary: Change the parameter on the command line from

    oracle.help.tools.index.Indexer

    to

    oracle.help.tools.index.JapaneseIndexer

    Thanks,
    Ken
  • 308977
    308977 Member Posts: 2
    Yes, I use the Japanese Indexer.

    Maybe it's simply because Japanese is one of the hardest languages to create an index file for!

    Thanks for your help anyway!
This discussion has been closed.